Autonomous Unmanned Systems Research Articles

Collaborative human-autonomy semantic sensing through structured POMDP planning

Autonomous unmanned systems and robots must be able to actively leverage all available information sources — including imprecise but readily available semantic observations provided by human collaborators. This work develops and validates a novel active collaborative human–machine sensing solution for robotic information gathering and optimal decision making problems, with an example implementation of a dynamic target search scenario. Our approach uses continuous partially observable Markov decision process (CPOMDP) planning to generate vehicle trajectories that optimally exploit imperfect detection data from onboard sensors, as well as semantic natural language observations that can be specifically requested from human sensors. The key innovations are a method for the inclusion of a human querying/sensing model in a CPOMDP based autonomous decision making process, as well as a scalable hierarchical Gaussian mixture model formulation for efficiently solving CPOMDPs with semantic observations in continuous dynamic state spaces. Unlike previous state-of-the-art approaches this allows planning in large, complex, highly segmented environments. Our solution is demonstrated and validated with a real human–robot team engaged in dynamic indoor target search and capture scenarios on a custom testbed.

Sliding mode approach for formation control of perturbed second-order autonomous unmanned systems

Sliding mode approach for formation control of perturbed second-order autonomous unmanned systems

Network-based leader-following formation control of second-order autonomous unmanned systems

This article investigates two kinds of leader-following formation control problems of second-order autonomous unmanned systems under directed topology, that are, the cases with constant velocity and time-varying velocity for the leader. First, the kinematical equation of autonomous unmanned systems is established according to some real scenarios. Second, by assuming the velocity of the leader is constant, a sufficient condition ensuring the formation control of autonomous unmanned systems is derived and some control parameters are suitably designed. Third, it is assumed that the velocity of leader is time-varying, and the formation controller with two parts is designed. The first part is intended to form the desired formation of all followers and the second part aims to reach the consensus of velocity. Some simulation examples are finally presented to verify the effectiveness of the obtained theoretical results.

Switched safe tracking control design for unmanned autonomous helicopter with disturbances

In this paper, the switched safe tracking control scheme is investigated for the attitude and altitude system of a medium-scale unmanned autonomous helicopter with output constraints and unknown external disturbances. To keep the attitude angles and altitude within the desired constrained range, an output boundary protection approach is adopted to generate an output constrained trajectory which is piecewise differentiable. The disturbance observer-based control method is employed to handle the unknown external disturbances of the system. Because of the piecewise differentiability of the output constrained trajectory, the closed-loop error system with the safe tracking controller can be seen as a switched system with jump dynamics. The multiple Lyapunov function method is adopted to guarantee the tracking performance with designed average dwell time. Simulation results of an example are provided to illustrate the effectiveness of the proposed control scheme for the unmanned autonomous helicopter system.

Robust TOA-Based UAS Navigation under Model Mismatch in GNSS-Denied Harsh Environments

Global Navigation Satellite Systems (GNSS) is the technology of choice for outdoor positioning purposes but has many limitations when used in safety-critical applications such Intelligent Transportation Systems (ITS) and Unmanned Autonomous Systems (UAS). Namely, its performance clearly degrades in harsh propagation conditions and is not reliable due to possible attacks or interference. Moreover, GNSS signals may not be available in the so-called GNSS-denied environments, such as deep urban canyons or indoors, and standard GNSS architectures do not provide the precision needed in ITS. Among the different alternatives, cellular signals (LTE/5G) may provide coverage in constrained urban environments and Ultra-Wideband (UWB) ranging is a promising solution to achieve high positioning accuracy. The key points impacting any time-of-arrival (TOA)-based navigation system are (i) the transmitters’ geometry, (ii) a perfectly known transmitters’ position, and (iii) the environment. In this contribution, we analyze the performance loss of alternative TOA-based navigation systems in real-life applications where we may have both transmitters’ position mismatch, harsh propagation environments, and GNSS-denied conditions. In addition, we propose new robust filtering methods able to cope with both effects up to a certain extent. Illustrative results in realistic scenarios are provided to support the discussion and show the performance improvement brought by the new methodologies with respect to the state-of-the-art.

Leader-following consensus tracking of autonomous unmanned systems with fixed/switching topologies: an event-triggered approach

In this paper, the leader-following consensus tracking of autonomous unmanned systems with fixed/switching topologies is investigated by event-triggered approach. Based on state estimates of each follower and its neighbours, a novel event-triggered strategy that can greatly reduce the communication frequency is proposed. On the basis of M-matrix theory, an event-triggered function is designed by suitably constructing a valid Lyapunov function. The communication likes are established when the event-triggered function is satisfied. Furthermore, suppose that each topology contains a directed spanning tree, it is shown that the consensus tracking of autonomous unmanned systems can be achieved under some conditions and the dwell time is bounded by a positive threshold. Zeno behaviour is proved to be excluded in systems. Two examples are given to demonstrate the validity of the proposed approach.

Coupling Traditional and Emergent Technologies for Improved Coastal Zone Mapping

The shallow, turbid water of the Delaware Bay Estuary is the second most navigated waterway in the USA behind the Mississippi River and experiences tropical and extratropical cyclones from June through April bringing high winds, storm surge, precipitation, and coastal change. Mapping coastal areas within the Delaware Bay is particularly difficult due to inherent environmental factors such as low underwater visibility (less than 1 m), rapid tidal current (greater than 1 m/s), changeable weather, and strong mid-latitude winds. This study utilized four sonar systems, two manned vessels, two autonomous vessels, three real-time kinematics global positioning systems (RTK GPS), two unmanned autonomous aerial systems, and satellite imagery to quantify subaerial and subaqueous volume and feature changes following storm events. The study site was located at Broadkill Beach, Delaware, a microtidal beach inside the Delaware estuary and outside the National Coastal Mapping Program survey area. Six storms (four tropical and two extratropical cyclones) from 2012 to 2018 were surveyed pre- and post-storm. Key insights in sensors and platforms proved unmanned aerial systems (UAS) orthomosaic resolution to be superior for mapping shoreline, profile, and dune toe changes, swash zone features such as cusps, wrack lines, and infrastructure damage to beach crossovers. Subaqueous platforms performed well for most storm events with RMSE of 0.23 m or less. This uncertainty, including subaerial uncertainty (max = 0.016 m), can be reduced with precise GPS systems and ground truthing of elevations through additional sensors like RTK GPS and subbottom profilers. No clear pattern emerged between accretion and erosion for tropical or extratropical cyclones (TC or ETC). Low-energy beaches are rarely studied and therefore open-coast relationships are transplanted to these environments. Storm responses of the site did not match expected traditional cross-shore transport models as beach orientation and storm wind direction most likely induced longshore transport out of the system. UAS orthomosaics were critical in identifying wrack lines and debris which altered subaerial transport patterns at this low-energy site. This work highlights the importance of irregular, small-scale features such as relic jetties, wrack lines, and local wind waves in estuarine sites when traditional open-coast theory would diminish these processes. More importance should be placed on the use of high-resolution imagery and elevation data over widely spaced profiles when monitoring and managing low-energy sites experiencing episodic storms such as TC and ETC to accurately assess processes, relationships, and morphologic response.

Binary GAN based Approach for Unsupervised Loop Closure Detection in Autonomous Unmanned Systems

Abstract Inspired by generative adversarial network (GAN), we propose a novel unsupervised approach for loop closure detection in autonomous unmanned systems. A binary GAN model dedicated to mobile application scenarios is designed to obtain binary feature descriptors, which are further incorporated into the most commonly used Bag of Visual Words (BoVW) model for loop closure detection. Compared with those hand-crafted features like SIFT and ORB, the performance of loop closure detection in complex environments with strong viewpoint and condition changes can be greatly improved. Compared with existing supervised approach based on convolutional neural network like AlexNet and AMOSNet, the cost-expensive task of supervised data annotation is totally avoided, which make the proposed approach more practical.

ПРИНЦИПИ РОЗРОБКИ МОДЕЛЬНОГО РЯДУ БЕЗПІЛОТНИХ АПАРАТІВ ДЛЯ ОБСЛУГОВУВАННЯ ПІДПРИЄМСТВ АГРОПРОМИСЛОВОГО СЕКТОРУ

The article discusses the issues of promising directions for improving monitoring tools and methods (information transfer to the operator in real time, thanks to video surveillance, fires are searched, survey of the territory), fire extinguishing (the arrival of a fire drone at the place of call at specified coordinates; fire identification by means of observation and control ; starting the fire extinguishing module; position stabilization when supplying fire extinguishing agents.), processing of fields (remote monitoring and control of accurate data on the topography of fields with all information about the unused area within the field using surveillance cameras; precision farming – using a drone with a combination of multispectral sensors, software products allows you to determine the need for plants in nitrogen fertilizers and other micro- and macronutrients, watering the field, analyzing the quality of the performed sowing work and even planning the harvest.). Analyzed digital methods of monitoring the airfield infrastructure. The possibilities of using drones in wide systems of their application and revealing their advantages are considered. Possible problems and their possible solutions are identified. The listed distribution companies in Ukraine that import drones such as: «Smart Trading (DJI, Parrot)», «MTI (DJI, Swell pro)», «ERC (Power Egg)», «Drone UA (DJI, Parrot, Drone Deploy, Mica Sense, Pix4D, Wingtra )», «L-trade (Leica Geosystems)», «Fly Technology (DJI, Yuneec)», «Quadro copter or DJI-kiev.com (DJI )». The directions of further research are determined and the categories of normative regulation of the use of unmanned vehicles are determined (remotely piloted aircraft systems; unmanned autonomous systems; legal use of remotely piloted aircraft systems). Based on the analysis of trends in the use of unmanned aerial vehicles and the directions of their further development, the conclusions are given, which states that the use of unmanned aerial vehicles is promising for firefighting, monitoring and field treatment. Their implementation is developing rapidly and they occupy a worthy place in aerial photography processes.

Acoustic camera modeling and compensation technique based on SONAR equation

Acoustic camera is a type of ultrasonic equipment used to detect or classify nearby objects located in water. The distortion of SONAR images acquired from conventional acoustic camera equipment inevitably occurs depending on the platform speed and operating frequency. This has not been huge concern because the platform speeds were low. However, more recently, a high accuracy of acoustic camera images has become essential in accordance with the development of unmanned autonomous systems. Additionally, a high speed of operating platform is also required. Therefore, there is a need for the study of image correction to reduce the distortion of SONAR images acquired from the equipment. In this study, we simulate virtual acoustic camera SONAR images by performing numerical modeling based on the SONAR equation to improve the accuracy of acoustic camera images. To verify the validity of the simulated images, we use the literature data and water-tank experiment data. Finally, we conduct the image compensation both in the direction of motion relative to the platform speed, and in the transverse direction relative to the platform speed and operating frequency.Acoustic camera is a type of ultrasonic equipment used to detect or classify nearby objects located in water. The distortion of SONAR images acquired from conventional acoustic camera equipment inevitably occurs depending on the platform speed and operating frequency. This has not been huge concern because the platform speeds were low. However, more recently, a high accuracy of acoustic camera images has become essential in accordance with the development of unmanned autonomous systems. Additionally, a high speed of operating platform is also required. Therefore, there is a need for the study of image correction to reduce the distortion of SONAR images acquired from the equipment. In this study, we simulate virtual acoustic camera SONAR images by performing numerical modeling based on the SONAR equation to improve the accuracy of acoustic camera images. To verify the validity of the simulated images, we use the literature data and water-tank experiment data. Finally, we conduct the image compensation both...

A Case-Based Economic Assessment of Robotics Employment in Precision Arable Farming

The need to intensify agriculture to meet increasing nutritional needs, in combination with the evolution of unmanned autonomous systems has led to the development of a series of “smart” farming technologies that are expected to replace or complement conventional machinery and human labor. This paper proposes a preliminary methodology for the economic analysis of the employment of robotic systems in arable farming. This methodology is based on the basic processes for estimating the use cost for agricultural machinery. However, for the case of robotic systems, no average norms for the majority of the operational parameters are available. Here, we propose a novel estimation process for these parameters in the case of robotic systems. As a case study, the operation of light cultivation has been selected due the technological readiness for this type of operation.

Quantum Network of Cooperative Unmanned Autonomous Systems

A quantum network may be realized by the entanglement of particles communicated by qubits between quantum computers, where the entangled photons of light are transferred for communication purposes. This technology has been proven to be feasible experimentally through free-space distribution of entangled photon pairs. Sending photons of light through nonlinear crystals produces correlated photon pairs, by splitting each photon into two half particles with each particle having the same level of energy, which results in entangled pairs. This entanglement is represented by photons, having both either horizontal or vertical polarization. This paper investigates collaborative robotic tasks of unmanned systems in a network where the agents are entangled. For instance, a leader robot sends two identical photons (e.g. with vertical polarization) to two follower robots/autonomous vehicles to communicate information about various tasks such as swarm, formation, trajectory tracking, path following and collaborative tasks. The potential advantages of quantum cooperation of robotic agents is the speed of the process, the ability to achieve security with immunity against cyberattacks, and fault tolerance, through entanglement. If a Quantum Network is implemented in a robotic application, it would present an effective solution; for example, for a group of unmanned systems working securely together. An analytical basis of such systems is investigated in this paper, and the formulation of quantum cooperation of unmanned systems is presented and discussed. The concept of experimental quantum entanglement, as well as quantum cryptography (QC), for robotics applications is presented.

An Intelligent Gain based Ant Colony Optimisation Method for Path Planning of Unmanned Ground Vehicles

In many of the military applications, path planning is one of the crucial decision-making strategies in an unmanned autonomous system. Many intelligent approaches to pathfinding and generation have been derived in the past decade. Energy reduction (cost and time) during pathfinding is a herculean task. Optimal path planning not only means the shortest path but also finding one in the minimised cost and time. In this paper, an intelligent gain based ant colony optimisation and gain based green-ant (GG-Ant) have been proposed with an efficient path and least computation time than the recent state-of-the-art intelligent techniques. Simulation has been done under different conditions and results outperform the existing ant colony optimisation (ACO) and green-ant techniques with respect to the computation time and path length.

Modeling of upset sensor operation for autonomous unmanned systems applications

PurposeThe purpose of this paper is to develop and implement a general sensor model under normal and abnormal operational conditions including nine functional categories (FCs) to provide additional tools for the design, testing and evaluation of unmanned aerial systems within the West Virginia University unmanned air systems (UAS) simulation environment.Design/methodology/approachThe characteristics under normal and abnormal operation of various types of sensors typically used for UAS control are classified within nine FCs. A general and comprehensive framework for sensor modeling is defined as a sequential alteration of the exact value of the measurand corresponding to each FC. Simple mathematical and logical algorithms are used in this process. Each FC is characterized by several parameters, which may be maintained constant or may vary during simulation. The user has maximum flexibility in selecting values for the parameters within and outside sensor design ranges. These values can be set to change at pre-defined moments, such that permanent and intermittent scenarios can be simulated. Sensor outputs are integrated with the autonomous flight simulation allowing for evaluation and analysis of control laws.FindingsThe developed sensor model can provide the desirable levels of realism necessary for assessing UAS behavior and dynamic response under sensor failure conditions, as well as evaluating the performance of autonomous flight control laws.Research limitations/implicationsDue to its generality and flexibility, the proposed sensor model allows detailed insight into the dynamic implications of sensor functionality on the performance of control algorithms. It may open new directions for investigating the synergistic interactions between sensors and control systems and lead to improvements in both areas.Practical implicationsThe implementation of the proposed sensor model provides a valuable and flexible simulation tool that can support system design for safety purposes. Specifically, it can address directly the analysis and design of fault tolerant flight control laws for autonomous UASs. The proposed model can be easily customized to be used for different complex dynamic systems.Originality/valueIn this paper, information on sensor functionality is fused and organized to develop a general and comprehensive framework for sensor modeling at normal and abnormal operational conditions. The implementation of the proposed approach enhances significantly the capability of the UAS simulation environment to address important issues related to the design of control laws with high performance and desirable robustness for safety purposes.

US Defense Innovation and Artificial Intelligence

The United States seeks conventional military overmatch in the strategic competition with China and Russia through efforts to develop advanced technology for military application and new concept of operations as well as pursue organizational adaptation and human talent pool development. The current article will trace the development of US defense innovation efforts from the Obama administration to the Trump administration and focus on the technological dimension of defense innovation by illustrating ways through which the U.S. Department of Defense is attempting to apply “narrow” artificial intelligence for military purposes such as situational awareness, cyber, military logistics, command and control, and swarm technology and tactics for autonomous unmanned systems. The article will conclude by making two caveats regarding the actual introduction of artificial intelligence into the US battle network, and very briefly point to implications for US allies.

Resilient Design and Operation of Cyber Physical Systems with Emphasis on Unmanned Autonomous Systems

Autonomy and autonomous systems are occupying central stage in the research community, as autonomous vehicles are proliferating and their utility in all aspects of the military and civilian domains are increasing exponentially from one year to the next. The development and application of resiliency and safety technologies to autonomous systems is, unfortunately, not keeping pace with their growth rate. Several factors impede the deployment and adoption of autonomous systems. Among them is the absence of an adequately high level of autonomy that can be relied upon, significant challenges in the area of human-machine interface requiring significant human intervention to operate and interpret sensor data, the need for emerging machine learning technologies and, most importantly, the resilient design and operation of complex systems to assure their safety, reliability and availability when executing missions in unstructured and cluttered environments. Recent advances in resiliency and safety of complex engineered systems have focused on methods/tools to tradeoff system performance for increased time to failure aiming at mission completion or trial and error methods to arrive at a suboptimal policy for system self-organization in the presence of a failure mode. This paper introduces a novel framework for the resilient design and operation of such complex systems via self-organization and control reconfiguration strategies that avoid empirical trial and error techniques and may be implemented and perform in real time on-platform. The main theme is summarized as: “a healthy and resilient system is a safe system”. To accomplish this objective, we introduce an integrated and rigorous approach to resilient design while safety considerations ascertain that the targeted system is contained within a safe envelope. A resilient system is robustly and flexibly monitoring its internal and external environment, it can detect and anticipate disturbances that may affect its operational integrity and take appropriate action to compensate for the disturbance. Resilience enhances safety while improving risk factors and assures that vehicles subjected to extreme disturbances remain within their safe envelope. The enabling technologies begin with graph spectral and epidemic spreading modeling tools to represent the system behaviors under normal and faulty conditions; a Markov Decision Process is the basic self-organization module. We are introducing a novel approach to fault-tolerance by considering the impacts of severe fault modes on system performance as inputs to a Reinforcement Learning (RL) strategy that trades off system performance with control activity in order to extend the Remaining Useful Life (RUL) of the unmanned system. Performance metrics are defined and assist in the algorithmic developments and their validation. We pursue an integrated and verifiable methodology to safety assurance that enables the evaluation of the effectiveness of risk management strategies. Several unmanned autonomous systems are used for demonstration purposes.

A Self-Organization Strategy for Unmanned Autonomous Systems

Complex systems are constructed from multiple subsystemsand components with each serving incremental tasks, wherethe “emergent” system behavior cannot be deduced from thebehaviors of the individual parts. The key requirement ofcomplex systems is the ability to compensate for unforeseenand extreme disturbances, so it is important to design acontrol method that ensures acceptable level of systemresilience throughout its operation. Therefore, detailed andaccurate knowledge of system behaviors is paramount for thedesign of complex system control strategies. This paperpresents a self-organizing control strategy that incorporatesboth situational awareness and failure impact compensationfor a resilient unmanned autonomous system.

Resilient-based Control Reconfiguration of Autonomous Systems

This paper introduces a design methodology for resilientbasedcontrol reconfiguration of Unmanned AutonomousSystems (UAS) when extreme disturbances, such as a largelygrowing fault or component failure mode occur. It isdocumented that more than 40% of Class air mishaps areattributed to Unmanned Aerial Vehicles (UAVs). There is anurgent need to improve the operational integrity, resilienceand reliability of such critical assets. An optimal controlapproach with Differential Dynamic Programming (DDP)and Model Predictive Control (MPC) is introduced in thispaper as a means for control authority redistribution andreconfiguration; therefore, the system continues performingits mission while compensating for the impact of the extremedisturbances. Prognostic knowledge is considered in aquadratic cost function of the optimal control problem as asoft constraint. A trade-off parameter is introduced betweenthe prognostic constraint and the terminal cost. Anautonomous ground operable under-actuated hovercraft isused to demonstrate the efficacy of the proposedreconfiguration strategy, and it is extendable to other cyberphysical systems.

Object-oriented and multi-scale target classification and recognition based on hierarchical ensemble learning

Target classification and recognition (TCR) of high resolution remote-sensing image is the important ability for earth observation system and unmanned autonomous system. It is difficult to improve the precision of TCR because of different imaging mechanism. In this paper, we propose a brain-inspired computing model for TCR using cognitive computing and deep learning. Accordingly, we have built an ensemble learning algorithm based on deep spiking convolutional neural network and hierarchical latent Dirichlet allocation. The hierarchical features were extracted from remote-sensing image. Then a TCR algorithm for small sample sizes and complex target was designed, which uses the incremental and reinforcement learning based on object-oriented and multi-scale data argumentation. Experimental results demonstrate that our algorithm has state-of-the-art performance on public data sets of optical remote-sensing image and synthetic aperture image. The model proposed can provide reference to explore an essential significance in brain-inspired intelligence, and has significant value in military and civil affairs.

Spatial Root Zone Soil Water Content Estimation in Agricultural Lands Using Bayesian-Based Artificial Neural Networks and High- Resolution Visual, NIR, and Thermal Imagery

Soil moisture is an important parameter in irrigation scheduling and application. Knowledge of root zone volumetric water content can support decisions for more efficient irrigation management by enabling estimation of required water application rates at appropriate temporal and spatial scales. The study presented here proposes a data mining approach that combines known field conditions with remote sensing observations to provide probabilistic estimates of root zone soil moisture at three different depths in the root zone soil profile. Bayesian data mining algorithms were tested and calibrated to combine the remotely sensed spatial information with field measurements. The remote sensing platform used in this study, called AggieAir™, is an unmanned autonomous aerial system developed by the Utah Water Research Laboratory at Utah State University. Finally, a trapezoidal integration method was used to estimate volumetric water content (VWC) in the root zone using the results of the modelling approach at three different depths in the root zone soil profile. The results show the model could estimate the root zone VWC with good accuracy (RMSE =0.05, MAE = 0.04, r = 0.97, e = 0.92, R2 = 0.94). Copyright © 2017 John Wiley & Sons, Ltd.